Agent Influence and Nested Other-Agent Behaviour

Stit theory, based on the notion of an agent's capacity for seeing to it that something holds, provides a rich semantic framework for manipulating agent ability. Its potential for representing nested statements readily draws stit theory onto uncomfortable ground with syntactically well formed constructs where one agent sees to it that another agent sees to it that something holds. This is uncomfortable because in systems of simple reactive agent it involves compromising the agency of the other agent.We introduce an alternative reading for stit using the notion of influence and this does not result in compromised agency. After introducing influence we briefly outline some of its logical properties and discuss results of exploratory experiments.